Temporal lobe epilepsy is common and difficult to treat. Our long-range research goal is to help reveal mechanisms of temporal lobe epilepsy and develop anti-epileptogenic strategies. During the last funding period we discovered in a rat model of temporal lobe epilepsy that initial loss of interneurons reduces numbers of GABAergic synapses with granule cells, but over time surviving interneurons sprout axons and develop excessive synapses. However, in epileptic dentate gyrus, although GABAergic synapses are abundant, at least some (for example, basket cell-to-granule cell synapses) are dysfunctional. During the next funding period, we propose 3 specific aims. Aim 1 is to test whether activation of mTOR signaling pathway contributes to GABAergic axon sprouting after epileptogenic injuries. We will use GIN mice, which express green fluorescent protein (GFP) in a subset of somatostatin interneurons. Extent of axon sprouting will be measured by stereological analysis of GFP-immunoreactive axons, biocytin-labeling of axons of individual GFP-positive interneurons, and probability of recording GFP-interneuron-to-granule cell unitary IPSCs (uIPSCs). Aim 1 will reveal whether or not mTOR signaling pathway contributes to epilepsy-related GABAergic axon sprouting. If not, based on previous findings, it would suggest mTOR pathway activation may be specific for excitatory mossy fiber sprouting. If so, it will establish a novel and innovative method for future studies to manipulate extent of GABAergic synaptogenesis and test whether it promotes or inhibits epileptogenesis. Aim 2 is to identify mechanisms underlying increased failure rate of basket cell-to-granule cell uIPSCs after epileptogenic injuries. We will use variance-mean analysis to estimate average release probabilities and minimum numbers of release sites by recording basket cell-to-granule cell uIPSCs in slices from control and epileptic pilocarpine- treated rats. Serial-section electron microscopy will be used to measure numbers of docked vesicles at GABAergic synapses with granule cell somata in control and epileptic rats. These experiments will more precisely identify why basket cell-to-granule cell synapses are more likely to fail in epileptic tissue, which will help future attempts restore normal function at these critical synapses. Aim 3 is to test whether hilar somatostatin interneuron-to-granule cell synapses are dysfunctional like basket cell-to-granule cell synapses after epileptogenic injuries. Approaches of Aim 2 will be used. Together, proposed experiments will advance understanding of how inhibitory synaptic transmission is affected in temporal lobe epilepsy. Ultimately, we expect data generated will help advance the long-term goal of developing anti-epileptogenic treatments.